Leakage in files metadata

Pri-matrix Factorization
1

Dear Organizers,

It seems that there is a leakage in the data you’ve provided for this competition.
I’ve built a simple experiment to check this out:

  • downloaded micro dataset
  • extracted time of last file modification and file size
  • used simple multiclass classifier with these two features
    Such model gave me a score 0.061752 on the leaderboard. For comparison, a sample submission with average class probabilities gives 0.090614 on the leaderboard.

Let us know what do you think on this issue.

Thanks,
Oleg

2

It probably can be not a leakage, but the feature of video encoder. Static videos basically has lower size. Videos with ‘blank’ label have larger probability to have smaller file size.

3

Yes, you’re right. I’ve also tried out to build a model with last file modification feature only and it has higher performance comparing to baseline with average probabilities as well.